PT832207E - METHOD FOR IDENTIFYING BIOLOGICALLY ACTIVE PEPTIDES AND NUCLEIC ACIDS - Google Patents

Abstract

Biologically active peptides and nucleic acids are identified by a method comprising the following steps: (a) production of a pool of appropriate vectors each containing totally or partly random DNA sequences, (b) efficient transduction of said vectors into a number of identical eukaryotic cells in such a way that a single ribonucleic acid and possibly peptide is expressed or a limited number of different random ribonucleic acids and peptides are expressed by each cell, (c) screening of said transduced cells to see whether some of them have changed a certain phenotypic trait, (d) selection and cloning of said changed cells, (e) isolation and sequencing of the vector DNA in said phenotypically changed cells, and (f) deducing the ribonucleic acid and peptide sequences from the DNA sequence. The peptide sequences may be introduced into or fused to a larger protein preferably an antibody molecule or a fragment thereof. This may be obtained by introducing the random DNA sequences into or fusing them to a DNA sequence encoding such larger protein.

Description

85 559 ΕΡ Ο 832 207 / ΡΤ

A method for identifying biologically active peptides and nucleic acids " The present invention relates to a novel method for identifying novel post-translated peptides and peptides as well as biologically active nucleic acids.

BACKGROUND OF THE INVENTION

During the last five years the technology for expression, testing and identification of millions of different random peptide sequences has evolved dramatically. Such peptide libraries can be used to identify new biologically active peptides and thus the technology has added to the field of drug development a new exciting and very promising time.

Known peptide library techniques can now be divided into two fundamentally different groups: random synthetic peptide libraries in which random peptides are chemically produced and random biosynthetic peptide libraries in which the random peptides are encoded by partial DNA sequences or totally random and subsequently synthesized by ribosomes.

Synthetic peptide libraries.

Synthetic peptide libraries containing millions of peptides can be produced by combinatorial peptide chemistry and can be synthesized in the soluble form (RA Houghten et al., Nature, 354: 84-86, 1991) or remain immobilized on the peptide resin beads (A. Furka et al., Int. J. Peptide Protein Res. 37: 487-493, 1991; KS Lam et al., Nature, 354: 82-84, 1991). Using either of these approaches, ligands of different receptors have been isolated. The advantage of the soluble peptide libraries compared to the solid phase immobilized peptide libraries is that the peptides 85 559 ΕΡ Ο 832 207 / ΡΤ 2

soluble compounds may be more stereochemically linked to the receptors in question. In the synthetic peptide library technique proposed by Furka et al. 1991 and Lam et al. 1991, respectively, the most important improvement was, on the other hand, the approach of having only one type of peptide sequence in each account (" An account - a peptide "). This allows a direct selection and possibly sequencing of the active putative peptide ligand in a single account using p. ex. degradation of Edman. Using this technology active peptides including peptides consisting of D-amino acids or other unnatural amino acids (B. Gissel et al., J. Peptide Science, Not published, 1995) may be identified.

Biosynthetic peptide libraries.

Expression bacteriophage vectors which may have peptides on the surface of the phage have been constructed (S. E. Cwirla et al., Proc Natl Acad Sci USA, 87: 6378-6382, 1990). Each peptide is encoded by a randomly mutated region of the phage genome such that sequencing the relevant DNA region of the bacteriophage that has been found to bind to a receptor will reveal the amino acid sequence of the peptide ligand. A phage containing a peptide ligand is detected by repeated panning procedures that enrich the phage population in a strong receptor binding phage (J.K. Scott, TiBS, 17: 241-245, 1992).

Bacteria were also used for expression of similar peptide libraries. The peptides may be fused to an exported protein, such as an antibody, which may be immobilized on a solid support. By screening the solid support with an appropriate soluble receptor the bacterial clone that produces the putative peptide ligand can be identified (M.G. Cull et al., Proc Natl Acad Sci USA 89: 1865-1869, 1992).

Another method described for preparing large pools of different possibly active compounds is through the use of libraries consisting of random ribonucleotides or deoxyribonucleotides, the so-called aptamers libraries. The aptamers are generated in E. coli from a plasmid vector containing random DNA. From these libraries, structures with biological activity (L.C. Bock et al., Nature, 355: 564-566, 1992) were identified. 85 559 ΕΡ Ο 832 207 / ΡΤ 3

OBJECT OF THE INVENTION

In order to utilize the prior art methods for identifying biologically active peptides and nucleic acids or their respective cellular target proteins, it is necessary to have a detailed knowledge about the molecular mechanisms involved in a particular biological process. If such mechanisms are known, it may subsequently be possible to develop antagonists or target agonists (receptors, enzymes, etc.) involved using said methods. The problem to be solved by the present invention is i.a. to overcome the need for such detailed knowledge.

SUMMARY OF THE INVENTION

According to the present invention the peptide sequences or ribonucleic acids are identified from biosynthetically expressed eukaryotic libraries containing millions of partially or totally random peptides and ribonucleic acids. In connection with this invention the term " peptide " is to be understood as also comprising a peptide sequence introduced into or fused with a larger protein (e.g., an antibody). The peptides and ribonucleic acids are synthesized by the cells from random DNA sequences that have been efficiently transduced into the cells. Some of the peptides or ribonucleic acids in the library will affect important biological functions in the cells that express them. Cells that change phenotype due to the presence of such substances can be isolated and their chemical structure can be subsequently clarified by sequencing the DNA encoding them. Such peptides may possibly be used therapeutically or as indicator compounds for future drug development or may be used to identify novel target proteins that are causing the change in the biological function of the cell. Such target proteins may optionally be used in the development of drugs through e.g. ex. conventional medical chemistry or synthetic peptide libraries.

Accordingly, the method of the invention comprises a method for identifying biologically active ribonucleic acids or peptides or for isolation of cellular ligands for biologically active ribonucleic acids or peptides, which comprises the steps of:

(A) producing a bank of suitable vectors each containing a DNA sequence to be examined, (b) efficiently transducing said vectors into several identical eukaryotic cells in such a way that each cell expresses a single ribonucleic acid and possibly a peptide encoded by the DNA sequence to be examined or a limited number of different ribonucleic acids and peptides encoded by the DNA sequences to be examined, (c) screening said transduced cells to see if any of them have altered one a certain phenotypic characteristic; and (d) selecting and cloning said altered cells, wherein the appropriate vector library in step (a) contains either fully or partially random DNA sequences selected from the group consisting of: (i) synthetic random DNA sequences obtainable through of conventional random oligonucleotide synthesis; ii) synthetic random DNA sequences in which the stop codons are absent and which encode random amino acid sequences with a balanced distribution of amino acids; iii) synthetic DNA sequences defined in (i) or (ii) including codons that allow for specific post-translational modifications of all peptides expressed or encoding anchor residues; iv) synthetic DNA sequences defined in (i), (ii) or (iii) fused with, or having inserted, coding sequences for purification markers that facilitate purification and identification of expressed peptides; and v) synthetic DNA sequences defined in (i), (ii), (iii) or (iv) inserted into or fused with the coding sequence of a protein;

And wherein (e) the vector DNA in the phenotypically altered cells is isolated and sequenced, and the sequences of the biologically active ribonucleic acids or peptides are deduced from the sequenced vector DNA; and / or (f) ribonucleic acids or biologically active peptides expressed on phenotypically altered cells are used directly for isolation of a ligand molecule for said ribonucleic acids or peptides.

BRIEF DESCRIPTION OF THE DRAWING Figure 1 is a schematic drawing of an expression vector of a standard retroviral peptide. The plasmid form of the vector (above) has a cytomegalovirus (CMV) promoter which directs expression of a retroviral RNA (at the center) with a structure (R, U5, PBS, ψ, PPT, U3 and R) of the Akv murine leukemia and a peptide translation cassette followed by an internal ribosomal entry site (IRES) of the EMC virus that directs the translation of a Neomycin (Neo) resistance gene. The vector provirus in the target cell (below) contains a regenerated retroviral long terminal repeat - LTR (U3, R, and U5). The CMV promoter sequence provides a unique marker for efficient initial PCR amplification and mutagenesis. The vector size without an inserted peptide expression cassette is 4.0 kb.

DESCRIPTION OF THE PREFERRED EMBODIMENTS

An intrinsic requirement of all currently known peptide library techniques is the need for detailed knowledge about the mechanisms by which a given receptor or enzyme regulates a particular phenotypic characteristic of the cell. This receptor or enzyme must furthermore be available in a relatively pure form before a ligand can be selected in any of the two types of peptide libraries. Where potential peptidic ligands are identified, functional tests have to be carried out to determine whether the

ligands exerts antagonistic or agonist effects on the desired cellular phenotypic characteristic. The method according to the present invention overcomes this major problem. By the present method it is thus not necessary to know the chain of mechanisms, receptors, signaling pathways, enzymes, etc. which generate the phenomenon inside or on the surface of the cell since it is the resultant biological effect or phenotypic characteristic that is being researched. This is achieved by the following steps: (a) producing a bank of appropriate vectors each containing fully or partially random DNA sequences, (b) efficiently transducing said vectors into several identical eukaryotic cells, e.g. ex. such that only a single species of ribonucleic acid and peptide (" a cell - a ribonucleic acid or peptide ") is expressed, or optionally a limited number of different ribonucleic acids and random peptides are expressed per cell, (d) selecting and cloning said altered cells, (e) isolating and sequencing the vector DNA in said phenotypically altered cells, and (f) RNA and peptide sequences from the DNA sequence. The RNA or peptide encoded by the isolated DNA sequences may be the cause of said phenotypic changes of the cell and may therefore have biological activity.

The introduced peptides are expressed e.g. ex. in the cytoplasm of biologically interesting cells, which before that were totally identical, using p. ex. the retroviral vector systems described in Example 1. These are expressed from a pool of vectors containing random DNA sequences that were constructed e.g. ex. as described in Example 1. Using a suitable ratio between infectious retroviruses and uninfected cells only a single copy of DNA derived from the pool of retroviral vectors is introduced into each cell. The main advantage of this concept of " a cell - a ribonucleic acid or peptide " is that cells undergoing phenotypic alteration upon introduction of peptides can be isolated by cloning and selection methods, and that the active peptide which causes the phenotypic alteration can be subsequently identified. This is achieved by isolating the DNA fragment encoding the peptide, e.g. ex. through Polymerase Chain Reaction (PCR) technology, and subsequently

85 559 ΕΡ Ο 832 207 / ΡΤ 7 identifying the DNA sequence.

During the initial screening procedure a larger number of retroviral vectors can be introduced into each cell which allows each of the cells to express several different ribonucleic acids or peptides. When a phenotypically altered cell clone is subsequently isolated all retroviral DNAs in that specific clone can be isolated by PCR and the PCR product can be used for retransfection of packaging cells normally used for virus production. Retroviral vectors isolated from these packaging cells may subsequently be used for the transduction of novel biologically interesting cells using the concept " a cell - a ribonucleic acid or peptide ". Finally, after a second cloning procedure the active substance can be identified as described above. In addition, the biologically active ribonucleic acid, peptide or protein isolated by this method can be used as a drug development indicator compound or as an affinity ligand for the purpose of isolating and identifying the protein target responsible for the biological activity. Such target proteins are very useful tools in drug development. The use of small vectors (below 3 kb of DNA) has the major advantage of allowing mutagenesis and simple PCR amplification without the steps of binding and cloning. Another important advantage of using small vectors is that vectors in a target cell bank can be directly amplified by PCR and further transfected into packaging cells, thereby allowing multiple rounds of selection to eliminate delayed false positive or cell analysis contaminants. Direct analysis of the sequences of the derived random plasmid clones is used to ensure randomness of the expression library. The standard vector capable of expressing the random peptide library is shown schematically in Fig.

The cells found to have been phenotypically altered after the introduction of the random DNA sequences could alternatively have been altered as a consequence of interactions with the ribonucleic acid molecule transcribed from the introduced DNA, in analogy with the disclosed libraries consisting of ribonucleotides or random polymorphonucleotides, the so-called aptamers libraries. Such ribonucleotide acid molecules would therefore also have biological activity. In addition, the observed effect could also be due to the biological activity of carbohydrate moieties, or other post-translational modifications, in the peptides expressed in the cell. Using a suitable purification label in the library peptides, it would be possible in this case to purify them and to analyze the exact chemical structure of the post-translational modifications in question.

Since the efficiency of the non-viral methods normally used for stable gene transfer to mammalian cells is very low, it would not be possible to establish a peptide library of expression in mammalian cells by such methods. Additionally non-viral methods generally lead to multiple DNA integrations in the cellular genome in disagreement with the concept " a cell - a ribonucleic acid or peptide ". To achieve the required high efficiency transfer of a single copy of a gene a viral vector must be used. Recently, cDNA expression libraries have been described which were constructed using retroviral vectors. From such libraries were isolated cytokines and cell growth factors (AJM Murphy et al., Proc Natl Acad Sci USA, 84: 8277-8281, 1987, BY Wong et al., J. Viro., 68: 5523-31, 1994, JR Rayner et al., Moi. Cell BioL, 14: 880-887, 1994). Expression of well-defined peptides in transfected eukaryotic cells was also previously established, although not using retroviral vectors (MS Malnati et al., Nature, 357: 702-704, 1992, EO Long et al., J. Immunol., 153 : 1487-1494, 1994). A random peptide library has never been expressed in mammalian cells for the purpose of identifying biologically active peptides or ribonucleic acids. Immunology is an important biological field where the method according to the invention can be used. T cells recognize only fragments of protein antigens and only if they are bound to MHC molecules. Two types of MHC molecules - the class I and II molecules - display antigen fragments to T cells. The peptides presented by MHC class I molecules, which are on the surface of essentially all nucleated cells, are 8-9 amino acids in length and are generally derived from cell cytosol proteins.

85 559 ΕΡ Ο 832 207 / ΡΤ 9

These may be auto-proteins, viral proteins, peptides introduced into the cytosol by transfection or tumor antigens. It is of considerable interest to be able to identify such T cell peptides or epitopes, e.g. ex. in relation to the development of vaccines or in cancer immunotherapy. Identification of such fragments is, however, a very difficult, demanding and sometimes impossible task requiring large amounts of affinity purified MHC molecules derived from the tumor cell in question and iterative combinations of advanced mass spectroscopy, HPLC, and functional assays of T cells. In addition, most peptidic antigens can not be identified because of the presence of very small amounts of each of the peptides in the MHC molecules. In Example 2 it is demonstrated that the method according to the present invention can be used for identification of said T-cell epitopes.

Cell lines expressing biologically important surface molecules may also be transduced with a random peptide library according to the invention. An example of such molecules could be the costimulatory molecule B7, which is known to be important for the activation of T cells, or the family of selectin proteins known to be involved in the routing of inflammatory cells to inflamed tissue. Phenotype-shifting cells (e.g., Which regulate B7 positively or negatively) can be selected and cloned as described in Example 3. After isolation of the DNA transduced by PCR, new cells can be transduced with the DNA alone to confirm the Note. Subsequently, the peptide sequence can be deduced from the DNA sequence.

It has been described by others that specific monoclonal antibodies or F (ab) fragments can be expressed in the cytoplasm of a cell and perform a biological activity there (T.M. Werge et al., Febs Letters, 274: 193-198, 1990).

According to the present invention the total or partially random peptide sequence may also be introduced into the variable region of an antibody F (ab) fragment. Therefore, a library of F (ab) fragments containing random peptide sequences may be expressed in a cell clone in a way that each of them expresses a single antibody specificity. Subsequently, biologically altered cells are isolated and

And the intracellular F (ab) fragment identified can be used for purification of the target protein involved in the biological proteins. Such a target protein may subsequently be used for the development of drugs capable of modifying said target proteins. The invention is illustrated by the following examples: EXAMPLE 1

Construction of an intracellular eukaryotic peptide library A retroviral vector capable of expressing random peptide sequences is constructed. If the random DNA sequences used in the vector were produced using conventional random oligonucleotide synthesis a large number of termination codons would inevitably be introduced. Furthermore, due to the degeneracy of the genetic code the result would be an unbalanced distribution of the encoded amino acids. We avoid this by producing random DNA sequences by random codon synthesis: An appropriate amount of resin used for solid phase oligonucleotide synthesis (optionally already containing a DNA sequence corresponding to an appropriate vector cloning site) is divided into 20 different portions. In part 1 a conventional solid phase chemical synthesis of three bases corresponding to a codon encoding the amino acid alanine is carried out. In portion 2 a codon encoding cysteine is synthesized and so on. When each of the 20 moieties was coupled with the codons corresponding to each of the natural amino acids, all portions are mixed and re-divided into 20 equal-sized resin portions. The codon synthesis is then repeated again and the entire procedure is repeated until the desired random DNA sequence has been synthesized - p. ex. corresponding to 6-10 random amino acids. This can also be achieved by using blocked and protected trinucleotide phosphoramidites encoding the natural 20 amino acids in a random total oligonucleotide synthesis (J. Sondek et al., Proc Natl Acad Sci USA 89: 3581-5, 1992). Finally, other suitable cloning sites of the vector may be synthesized in all oligonucleotides before they are eventually cleaved from the resin.

A random synthetic oligonucleotide library can be used to generate a library of vectors with random sequences at appropriate positions either by restriction cleavage and binding, or preferably to avoid inefficient binding steps, by PCR mutagenesis. The procedures follow the principles of site-directed PCR mediated mutagenesis (S. Perrin et al., Nucl Acids Res. 18: 7433-38, 1990), but the methodology was adapted to deliver a complex mixture of products. In brief, the

random oligonucleotides having vector sequences flanking the random sequence are used as primers in a PCR reaction together with a single vector terminator primer. To maintain complexity, large amounts of template as well as vector and random primers are used, and product diversity is further ensured by assembling multiple independent PCR reactions. Subsequently, an overlapping PCR fragment containing the remaining segment of the vector is produced by standard PCR. Finally, this overlapping segment is ligated to the PCR fragments containing the random DNA using another unique set of terminal primers.

The DNA fragments produced by the Taq DNA polymerase enzyme may contain additional nucleotides in the 3 'DNA strand. These extra nucleotides will be deleterious to the combination of two overlapping terminal PCR products on one fragment. By addition of the Klenow DNA polymerase enzyme these nucleotides can be removed by the 3 'exonuclease activity -> 5 'of said enzyme, increasing the combining efficiency.

Alternatively the PCR product may be cleaved at the termini by restriction enzyme digestion whose recognition sequences were incorporated into the oligonucleotides used as primers for the PCR reaction. By using the temperature cycle method developed by (Lund et al., Nucl Acids Res., 24: 800-801, 1996) the efficiency of the binding reaction can be increased and the product of the linkage used for direct transfection of the packaging cells.

To maintain diversity, the linear vector generated by PCR is used directly for transfection of packaging cells, and the virions containing the template of different vectors are harvested under transient conditions. 12 85 559 ΕΡ Ο 832 207 / ΡΤ

Small bicistronic vectors from a single transcript containing a random peptide translation cassette followed by an internal ribosomal entry site (IRES) of the EMC virus to direct translation of a Neomycin (Neo) resistance gene. The Neo gene functions as a selection marker to allow determination of the titer and, if necessary, elimination of non-transduced cells. Other relevant vectors employ other selectable genes such as resistance to phleomycin and hygromycin B and have the peptide translation cassette after the IRES element. Alternatively, even smaller vectors, having only the peptide expression cassette and without a selection marker, may also be used. The use of small vectors (below 4 kb of DNA) has the major advantage of allowing simple PCR mutagenesis and amplification without binding and cloning steps. Another important advantage of using small vectors is that the vectors in a target cell bank can be amplified directly by PCR and again transfected into packaging cells, thereby allowing multiple rounds of selection to eliminate a delayed false positive or contaminating cells. Direct analysis of the sequences of the derived random plasmid clones is used to ensure randomness of the expression library. The standard vector capable of expressing the random peptide library is shown schematically in Fig.

To maintain the diversity of the vector bank, a high fraction of the vector RNA transcripts must be encapsidated into retroviral particles. By transfecting the packaging cells with a DNA construct expressing a tRNA matched to the corresponding PBS in the retroviral vector, we can increase the production of functional vector containing viral particles under transient conditions by 10-fold.

A new packaging cell line will be generated after a single transfection of a construct encoding all of the retroviral proteins. To reduce the risk of replication-competent virus creation and to obtain maximum expression, the vector will have the following simplified outline: promoter-transcribed gag-pol-IRES-phleomycin resistance gene-IRES-env-polyadenylation signal. One advantage of such a vector is that the 13,355,559 Ρ Ο 832 207 / gene resistance gene for phleomycin allows a selection for high expression and that the size after the cut of the vector transcript restricts the encapsidation of retroviral particles thereby decreasing the risk of virus creation competent for replication. The size limit for encapsidation of RNA transcripts is about 10 kb.

In addition to the traditional packaging cell line a semi-packaging cell line with a corresponding miniviral vector will be used. The semi-packaging cell line consists of vectors encoding two mutated gag-pol transcripts that complement one another. The use of two different gag-pol transcripts reduces the risk of creating wild-type viruses. Each cell in the semi-packaging cell line now contains all of the retroviral proteins required for the production of retroviral particles except the envelope proteins, which proteins are provided by the miniviral vector. This vector is a bicistronic vector with the following sketch: LTR-PBS-polypeptide-LTR random-packaging-peptide signal-IRES-env-tract. This vector will be able to transduce the semi-packaging cells since they do not produce envelope proteins prior to infection with the miniviral vector. Thus, the infection of the semi-packaging cell will not be restricted by receptor interference.

Restrictions may be placed on " randomness " of the peptide sequence, in order to limit the number of sequences available in the mammalian cell library and for the introduction of p. ex. N-glycosylation sites, or other post-translational modifications of all expressed peptides, if so desired. Purification markers - p. ex. poly-His or others - may also be included in the expressed peptides to facilitate purification and identification of the peptide itself. This is necessary for the identification of post-translational modifications, which are not obvious from the peptide sequence.

A population of eukaryotic cells is infected with the retrovirus which possesses genetic constructs containing random DNA sequences which encode a library of millions of random peptides. Initially an excess number of viruses may be used in comparison with eukaryotic cells. This leads to the expression of several different peptides at each f

85 559

The present invention relates to a eukaryotic cell. These cells may subsequently be screened as described below and the DNA may be isolated e.g. ex. by PCR and used for reinfection of other cells. If an appropriate ratio between the number of retroviruses containing the random and cell DNA sequences is chosen, each cell will be transduced with a different random DNA sequence (" a cell - a ribonucleic acid or peptide "). This optionally allows the identification of an active peptide. The peptide may optionally be targeted to different compartments in the cell by incorporating appropriate signal sequences into the translated sequences. EXAMPLE 2

Identification of T-cell epitopes through the use of mammalian intracellular expression libraries The interleukin-2 (IL-2) dependent cell line, CTLL-2, is transfected with the major Histocompatibility Class (MHC) class I molecule, Kb, of murine. Subsequently this cell line is infected with a peptide retrovirus library which was described in Example 1. This peptide library of CTLL-2 expresses a wide variety of random peptides and, if appropriate, retroviral peptide sequences associated with Kb are known to be important for Kb peptide binding, and a large library of K ° -linked peptides is shown on the surface of CTLL-2 cells. Kb restricted T-cell hybridomas are generated against an appropriate viral antigen using conventional cellular immunological technology (Current Protocols in Immunology, Eds. Coligan et al., NIH). Such hybridomas secrete IL-2 upon recognition of the antigen. A T-cell hybridoma, which recognizes an unknown Kb-linked T-cell viral epitope, is subsequently incubated with samples from the CTLL-2 Kb library. If the hybridoma recognizes a peptide, the CTLL-2 cell bearing the peptide will be stimulated to proliferate by the hybridoma secreted IL-2. In this way the CTLL-2 cell expressing the unknown T-cell viral epitope can be selected and cloned. The DNA sequence encoding the peptidic epitope

85 559 ΕΡ Ο 832 207 / ΡΤ 15 can be isolated from this clone using PCR technology followed by conventional DNA sequencing. This eventually leads to the identification of the unknown T cell viral epitope. EXAMPLE 3

Identification of biologically active peptides or ribonucleic acids that regulate the expression of proteins on the cell surface B7 immunoregulatory membrane molecule expressing cells are infected with the random peptide libraries constructed as described in Example 1. In this example, a library of eight- mers.

Using specific monoclonal antibodies, B7 expression is analyzed by conventional methods. Cells that regulate B7 positively or negatively can be selected positively, e.g. ex. using Fluorescence Activated Cell Separation, or negatively, e.g. ex. using appropriate antibodies in combination with complement lysis.

Cells showing changes in B7 expression are cloned by conventional means and DNA introduced through retroviral vectors is isolated using PCR and a set of specific retroviral primers. The peptide sequence or possibly the RNA corresponding to said DNA may be capable of modifying the expression of B7 and thus the activation of T cells. This can subsequently be tested in conventional T cell assays. EXAMPLE 4

Identification of an F (ab) fragment capable of modifying the B7 immunoregulatory molecule. A retroviral library is produced encoding the variable heavy chain (VH) gene fragments as well as the light variable (VL) gene of the immunoglobulin molecule. The gene region of both the gene fragments corresponding to the antigen binding site of the resulting F (ab) fragments also contains partially random gene sequences as described in Example 1. This will lead to a large number of peptide sequences at the antigen binding site of the F (ab) fragment. The retroviral vector library therefore encodes a large number of different F (ab) fragments with different antigen binding specifications.

This library is subsequently transduced into a cell clone such that each cell expresses a single F (ab) p fragment species. ex. in the cytoplasm. Phenotypically altered cells are subsequently cloned and the peptide sequence on the intracellular F (ab) fragment is identified as described in Example 1. This antibody can be subsequently produced on a large scale by conventional means and used for purification of protein affinity target cell responsible for the biological alteration of the cellular phenotype. This may e.g. ex. be made from lysates produced from the original unmodified cell clone.

Alternatively, the retroviral F (ab) library can be constructed using p. ex. a poly-His tag or other appropriate labels. In this way the antibody and the corresponding target can be isolated directly from the phenotypically altered cell by affinity chromatography. The isolated target can subsequently be identified through p. ex. N-terminal amino acid sequencing in combination with conventional cloning methodology. Such target proteins are drug targets very important for the identification of more drugs.

Lisbon, 13. DF7 2C-G0

By M & E BIOTECH A / S - THE OFFICIAL AGENT-

Claims (31)

A method for identifying biologically active ribonucleic acids or peptides or for isolation of cellular ligands for biologically active ribonucleic acids or peptides, which comprises the steps of: (a) producing of a bank of suitable vectors each containing a DNA sequence to be examined, (b) efficiently transducing said vectors into several identical eukaryotic cells such that each cell expresses a single ribonucleic acid and possibly a peptide encoded by the DNA sequence to be examined or a limited number of different ribonucleic acids and peptides encoded by DNA sequences to be examined, (c) screening said transduced cells to see if any of them have altered a particular phenotypic characteristic; and (d) selecting and cloning said cells, characterized in that the appropriate vector library in step (a) contains total or partially random DNA sequences selected from the group consisting of: (i) synthetic random DNA sequences obtainable through conventional synthesis of random oligonucleotides; ii) synthetic random DNA sequences in which the stop codons are absent and which encode random amino acid sequences with a balanced distribution of amino acids; iii) synthetic DNA sequences defined in (i) or (ii) including codons that allow for specific post-translational modifications of all expressed peptides or which encode anchor residues; iv) synthetic DNA sequences defined in (i), (ii) or (iii) fused to, or encoded within, 85 559 ΕΡ Ο 832 207 / ΡΤ 2/6 purification which facilitates the purification and identification of the expressed peptides; and v) synthetic DNA sequences defined in (i), (ii), (iii) or (iv) inserted into or fused with the coding sequence of a protein; and wherein (e) the vector DNA in the phenotypically altered cells is isolated and sequenced, and the sequences of the biologically active ribonucleic acids or peptides are deduced from the vector DNA sequenced; and / or (f) ribonucleic acids or biologically active peptides expressed on phenotypically altered cells are used directly for isolation of a ligand molecule for said ribonucleic acids or peptides. A method according to claim 1, wherein the identical eukaryotic cells are cells of a cell line. A method according to claim 1 or 2, wherein the peptide is a peptide sequence introduced into or fused to a protein. A method according to claim 3, wherein the protein is a. F (ab) fragment or an antibody molecule. A method according to any one of claims 1-4, wherein the synthetic DNA / oligonucleotide sequences have been produced by random codon synthesis, where defined DNA codons are synthesized in a random order. A method according to claims 1-4, in which the synthetic DNA / oligonucleotide sequences were produced by conventional random oligonucleotide synthesis. 85 559 ΕΡ Ο 832 207 / ΡΤ 3/6 A method according to any one of claims 1-6, wherein the random DNA sequences are introduced into the expression vector by site-directed PCR mediated mutagenesis thereby ensuring the complexity of the total or partially random DNA sequences. A method according to claim 7 in which the 3 '- * 5' exonuclease cut from the 3 'ends of the PCR product is used for optimum combining efficiencies of two PCR products. A method according to any one of claims 1-8 in which the random DNA sequences are introduced into the various eukaryotic cells such that only one DNA sequence is introduced into each cell, one cell expressing a ribonucleic acid and possibly a peptide, thus allowing a given sequence to be isolated and analyzed. A method according to any one of claims 1-9, wherein random DNA sequences are introduced into eukaryotic cells by use of appropriate viral vectors selected from retroviral vectors and vaccinia virus vectors. A method according to claim 10, in which the vector used is a retroviral vector. A method according to claim 10, in which the vector used is a vaccinia virus vector. A method according to claim 11, in which the retroviral vector contains a CMV promoter to replace the viral promoter in the 5 'LTR. A method according to any one of claims 10-13, in which random DNA sequences are produced as linear PCR products which are introduced directly into virus packaging cells by non-viral transfection methods. A method according to any one of claims 10-14, wherein Which is the viral DNA introduced into the cells is amplified directly through PCR and used for further transfection of packaging cells to produce viruses subsequently used for transduction of new cells with the aim of eliminating false positives and / or allow the concept " a cell - a ribonucleic acid or peptide ". A method according to any one of claims 10-15, wherein the viral titer of the retroviral packaging cell lines is increased by transient transfection with a functional tRNA gene corresponding to the PBS in the vector. A method according to any one of claims 10-16 in which a packaging cell line constructed from a vector expressing a single transcript is used which is translated into the three polyproteins / proteins, gag-pol, a drug resistance marker and env. A method according to any one of claims 10-17, in which a semi-packaging cell line is used with a corresponding miniviral vector which allows expression of the vector after transduction rather than after transfection of the cells. A method according to any one of claims 1-18, wherein the expressed peptides are modified after translation or include anchor residues. A method according to claim 19, wherein the post-translational modifications comprise the introduction of glycosylation sites. A method according to any one of claims 1-20 in which the biologically active peptide or expressed protein containing the random peptide also contains a purification marker that allows direct isolation of the biologically active protein as well as the target protein that causes the biological activity. A method according to any one of claims 1-21, in which also suitable signal peptides are encoded, other molecules of By the introduced DNA such that they are fused to the expressed random peptides or to the expressed proteins containing the random peptide sequences, allowing them to be directed into compartments defined cell phones. A method according to any one of claims 1-22, wherein the random DNA sequences are introduced into or fused with a DNA sequence encoding a protein expressed simultaneously with the random DNA sequences from the vectors of library. The method of claim 23, wherein the protein is selected from the group consisting of secreted proteins, intracellular proteins and membrane proteins e.g. ex. signal transduction molecules. The method of claim 23 or 24, wherein the protein is derived wholly or partially from the heavy and / or light chain of an antibody molecule. A method according to any one of claims 1-25, which is used for the identification of T-cell epitopes. A method according to any one of claims 1-25, which is used for the identification of biologically active peptides that regulate the expression of cell surface proteins. A method according to any one of the preceding claims, wherein the identical eukaryotic cells are mammalian cells. A method according to any one of the preceding claims, wherein the random DNA sequences encode 6-10 random amino acids. A method of drug development comprising the identification of biologically active ribonucleic acids or peptides or the isolation of cellular ligands therefrom according to the method of any one of claims 1-28. A method according to claim 30, wherein the identified biologically active ribonucleic acids or peptides serve as indicator compounds. BIOTECH A / S - THE OFFICIAL AGENT -